Best one step ahead prediction in hydrology

摘要: Recent papers suggest that stochastic one step ahead prediction is an important growth field in hydrology; two examples of this viewpoint are by Rao and Kashyap [1971] Carlson et al. [1970]. More specifically, it beginning to appear ‘best prediction’ be equated the quasi objective procedure developed Box Jenkins for other purposes. For uninitiated let me state a recipe can follow if wishes choose specific model from domain autoregressive integrated moving average models (ARIMA) (p, d, q). Even we set aside all technical problems relative selecting criteria, small sample biases, seasonal diurnal variability, what might best called type 2 error fitting problem, there still philosophical evaluation connected with applying approach short-term hydrologic forecasting.